Calendar

The accurate and timely prediction of solar eruptions is important for many space weather prediction tools and the Solar Orbiter mission. The aim of this study is to propose a new technique for the automated prediction of magnetic flux rope ejections in data driven NLFFF simulations hours in advance. We use a data-driven NLFFF model to describe the evolution of the 3D magnetic field of 8 active regions: 5 that produced an eruption and 3 where no eruption was observed. From the 3D magnetic field configuration, we determine a possible proxy for the loss of equilibrium of the magnetic flux rope based on the Lorentz force. Such proxy is significantly higher for the simulations of the eruptive active regions. For some cases, using a subset of the observed magnetograms, we ran a series of predictive simulations to test whether the time evolution of the proxy project forward in time can be used to predict the eruptions. We find that the identified proxy is useful in anticipating the magnetic flux rope ejection and that a meaningful prediction can be made up to 10 hours in advance. Although a number of issues need to be addressed for a fully operational application, this study presents an interesting solution for the prediction of CME onsets and future studies will address how to generalise the model such that it can be used.